Western blotting has moved from "is the band there?" to "how much protein is in the band?" Fluorescent detection is a big reason. Fluorescent labels give a signal that is stable over the imaging session, and several targets can be read on one membrane. That turns the blot imager into a measuring instrument, and it changes what the camera inside it has to do.
This post is for instrument makers designing or updating a fluorescent blot imager. It covers the camera questions that decide whether band intensities can be trusted, using Finger Lakes Instrumentation (FLI) Revelation cameras as examples. If your imager still uses a CCD, start with our post on replacing the CCD in gel and blot imagers.
Why fluorescent blots changed the camera brief
Chemiluminescence is an enzyme reaction. The light builds and fades, so the signal depends on timing as much as on protein. Fluorescence is steadier: excite the dye, collect the emission, and the result can be repeated.
That puts three demands on the detector. Its response has to stay linear over the range of band intensities on a real blot. It has to keep separate channels separate. And it has to measure the same band the same way wherever it sits on the membrane.
Linear range is the number that matters
Datasheets often quote dynamic range as full well divided by read noise. For the RV533 and RV571, the published typical values are a 50,000 e- full well and 1 e- read noise in rolling mode, which works out to about 50,000:1 in a single frame. That is a useful starting point, but those two figures may be measured at different gain settings, so the range you get at any one setting can be smaller.
For quantitation, the figure to measure is the linear range: the span over which signal tracks the amount of light within your tolerance. Measure it at the gain and readout settings your software will use, with a dilution series on a real membrane. Our post on acceptance testing shows how to check gain and linearity on arrival.
Saturation and exposure strategy
A saturated band is a lost measurement. Once the brightest pixels in a band reach full well, the integrated intensity stops rising with protein load, and the band looks weaker than it is relative to its neighbors.
Imager software should flag saturated pixels in every capture and refuse to report a value for a band that contains them. Auto-exposure should aim for the strongest band of interest, not the brightest spot on the membrane.
When strong and weak bands share a membrane, low read noise opens another option. Take several shorter exposures and add them. Read noise adds in quadrature, so ten frames at 1 e- carry about 3.2 e- of combined read noise, while each frame stays below saturation.
Fluorescent exposures are usually short compared with chemiluminescence, so dark current is rarely the limit. At the RV533 and RV571's typical 0.002 e-/pixel/s at -20 °C, a 60 s exposure collects about 0.12 e- of dark signal per pixel, well below the read noise.
Multiplexing: channels, filters and crosstalk
Multiplexing lets a lab read a target and a loading control, or two targets, on the same membrane with no stripping and reprobing. Each channel needs its own excitation and emission band, and the bands have to be far enough apart that one dye does not show up in another dye's channel.
Bleed-through is an optics problem more than a camera problem, but the camera sets the budget. A sensor with low noise across the visible spectrum lets you use narrower emission filters and still get a clean signal, which reduces crosstalk. Measure it during design by imaging single-dye blots in every channel.
Window choice matters at the spectrum's edges. FLI's F116 window transmits farther into the near-IR, and F101 transmits farther into the near-UV. Ask for the transmission charts, and decide early, because the window is chosen before the camera enters our clean-room build.
A blot imager with a handful of channels can often use a standard filter wheel, since a move of about 1 s is small next to the exposure. If your instrument steps through many channels, or runs blots back to back, the FLI High Speed Filter Wheel moves to the adjacent filter in 23 ms on the 6-position model and 30 ms on the 10-position model. See our post on fast filter switching.
Total protein normalization adds a channel
Normalization is where quantitative blots often go wrong. Eaton and colleagues showed that common loading controls such as actin and tubulin vary across tissues and disease models. In one mouse model, beta-actin was down by about 19% compared with healthy controls, enough to skew normalized results by about 20%. They found that normalizing to total protein, using stains such as Coomassie, was more robust (PLoS ONE, 2013).
For an imager maker, total protein normalization means one more image per blot, often under different illumination from the target channels. Plan the light source, filter position and exposure presets for it from the start, and make sure the software ties each total-protein image to its target images.
Uniformity across the membrane
The same band should read the same in a corner lane as in the center. Illumination falloff and lens vignetting break that, and the error is different in each channel.
Build a flat-field correction for every channel from a uniform fluorescent target, and repeat it when the optics or light source change. Dark frames come next. Revelation cameras do not use a mechanical shutter, so take darks with the excitation off in the closed cabinet, or with a blocked position on the filter wheel.
Format, optics and the cabinet
The RV533's 11.3 × 11.3 mm sensor suits 1-inch-class optics and square membranes. The RV571's 23.5 × 15.7 mm sensor covers wider fields or finer sampling with optics that can fill it. Both use 3.76 µm pixels, and binning on these Sony sensors is digital, so check sampling and noise together. Our post on pixel size and sampling walks through the math.
Revelation shares the electronics and sensors of our Aurora cameras in a thin body built for tight cabinets. It cools to about 50 °C below ambient on air, with liquid cooling or mounting to your housing as a heatsink as options.
Calibration, software and validation
Every Revelation camera includes our software development kit and support for integration into your OEM design, so exposure, gain, readout and saturation checks can live in your own application. Drivers and manuals are on our support page.
Validate with the same reference blots and dilution series your customers will trust: linear range per channel, crosstalk, flat-field residuals and repeatability across units.
Questions to settle before you choose
- How many fluorescent channels, and at which wavelengths?
- What linear range do your customers need on one membrane?
- How will your software detect and handle saturation?
- Will total protein normalization be imaged, and under what illumination?
- What field of view and optics does your cabinet use?
- How much space, airflow or coolant is available?
Talk to our sales team
FLI cameras are Made in USA, and our sensor chambers are assembled in our 1,000+ sq ft ISO Class 7 clean room. See how on our in-house manufacturing page.
Send our sales team your channel list, membrane size and optical layout, and we'll help you choose a camera, window and filter wheel for your imager. Request a quote or email sales@flicamera.com.
